9/04/2019

Objectives

Develop recommendations for underwater survey methods for monitoring zebra mussels in recently infested lakes:

  • Estimate density
  • Distribution within the lake
  • Quantify effects of treatments and impacts that depend on density
  • \(\Rightarrow\) requires comparisons over time and space
image: Naomi Blinick

image: Naomi Blinick

Why Focus on Newly Infested Lakes?

Existing survey designs


image: Naomi Blinick


  • Discovery in low densities (timed searches)
  • High-densities (quadrat surveys)


But what about designs for low to moderate densities?

Important Considerations

  • We can’t sample entire lakes!

    • must infer density after sampling only part of a lake
    • use random or systematic sampling

  • We may not detect all mussels in the area we sample

    • use methods that can correct for imperfect detection

Important Considerations

.

Our ability to detect mussels may depend on

  • water clarity, presence of plants, …
  • the diver

Year 1: Surveys Using Distance Sampling

Year 1 field crew

image: Naomi Blinick

image: Naomi Blinick

Transect sampling

Transect Sampling + Distance Sampling

Allows us to and account for imperfect detection.

Important assumptions in conventional distance sampling

  • Densities are uniform with respect to distance from the transect line (guaranteed with random placement of transects)
  • Detection on the transect line is perfect
  • Animals do not move before detection
  • Measurements are exact

Extra information yields detection estimates

Leca, J., N. Gunst, A. Rompis, G. Soma, I. G. A. Arta Putra, and I. N. Wandia (2013) Population Density and Abundance of Ebony Leaf Monkeys (Trachypithecus auratus) in West Bali National Park, Indonesia, Primate Conservation 26(1), 133-144.


Lake Sylvia

Lake Burgan

Estimating detection with two observers

  • First observer marks each detected cluster
  • Second observer looks for new clusters
  • New detections inform our estimate of detection probability

Importance of Accounting for Detection

image: Naomi Blinick

image: Naomi Blinick

Estimated density without detection \(0.08\) mussels/m\(^2\).

Estimated density without detection \(0.25\) \((0.07)\) mussels/m\(^2\) .


Correcting for imperfect detection lead to a 3-fold increase in our estimate of density!

Year 2: Comparing Multiple Survey Designs

Dive team

How do distance surveys compare to quadrat surveys?

By surveying smaller quadrats, we may be able to detect all mussels in the surveyed area.

When might quadrat designs be preferable?

image: Jake Ferguson

image: Jake Ferguson

Given a fixed amount of time, which method performs best?

We compared designs across densities


Results

Lessons from Year 2

  • Distance sampling is preferable at low to moderate densities

  • Quadrat sampling may be preferable at high densities (when?)

image: Aislyn Keyes

image: Aislyn Keyes

Generalizing these results

Ongoing Work

  • Using simulations (work with Dr. Katie St. Clair at Carleton College)



  • Analytic approach that shares similarities with models used to determine optimal foraging behaviors (Dr. Jake Ferguson, University of Hawaii)

Additional Products

Training video

Resources

Acknowledgements

Jake Ferguson

Michael McCartney

Naomi Blinick

Leslie Schroeder

Sarah Baker

Aislyn Keyes

Austin Hilding

Kylie Cattoor

Keegan Lund

Quadrat sampling error